A DNS Study of the Dispersion and Deposition of Nano- and Micro-Particles in a Turbulent Channel Flow

Abstract

Nano- and micro-particles dispersion and deposition in a dilute gas-solid turbulent flow in a channel were studied. A pseudo-spectral DNS code was used to solve the Navier-Stokes equations, and to generate the instantaneous turbulent velocity fluctuation field for the gas flow. Under the one-way coupling assumption, the gas flow carries the particles, but the influence of particles on the flow can be neglected. To provide an understanding of the transport behavior of particles of different sizes, 200,000 monodisperse point particles with Stokes numbers of 0.1, 1, 5, 25, and 125 were introduced with a random distribution in the channel. The corresponding Lagrangian particle equation of motion, including the Stokes drag, the gravity, and the lift forces, were solved, and the trajectories of the particles for the duration of 10,000 wall units were evaluated for dilute suspension. The trap boundary condition on the lower and upper walls of the channel was assumed, and the deposition rates of particles with different sizes were evaluated and recorded as a function of time. Ensemble and time averaging of the simulation results were performed, and the corresponding concentration profiles and the deposition velocities of particles were evaluated for various conditions. A series of simulations were performed, and the effects of wall roughness, lift force, and the gravity direction on the deposition rate were carefully examined. It was found that the surface roughness and the direction of gravity in conjunction with the lift force significantly affect the fine particle deposition rate and could improve the agreement of the DNS simulation with the available experimental data.